Reviewing “Replacing Darwin” – Part 2: Darwin Didn’t Know About Chromosomes!

Reviewing “Replacing Darwin” – Part 2: Darwin Didn’t Know About Chromosomes!


Here we go for round 2, let’s get straight into the next 2 chapters. The first focuses on inheritance while the second is about the genetic basis of traits.

Chapter 2 – The Secret of Life

Jeanson begins by reiterating what he said in the last chapter: species are defined by traits so the question of the origin of species is a question about the origin of traits. A key part of this question is understanding how traits behave over time – how they’re inherited.

He points out that if you track traits through human family trees, you notice that sometimes traits disappear and reappear – a grandmother with red hair has offspring without red hair, but those offspring have children that do have red hair. This kind of observation in isolation is described as a “paradox”, and he then goes on to claim that this paradox had profound implications for the question of how traits behave over time in Darwin’s day.

These paradoxes would raise a host of perplexing questions. Do traits form spontaneously? Can they be destroyed? Can they be changed? If so, how much can they be changed? Are traits blended? Particulate? Inherited as a whole? Separated into units? Independent? Interdependent?

Consider the ramifications of this uncertainty. If traits can appear and disappear, how could you trace species’ ancestry? What markers would you use to fill in the family tree? Furthermore, if all traits are rebuilt every generation, can any species become any other species? Do any constraints on change exist? Might a fish spontaneously spawn a spider? Without an answer to the mystery of heredity, the origin of species would be an enigma.

It’s true that, in 1859, Darwin (and everyone else) didn’t have a good understanding of the nature of heredity, but that didn’t stop him from observing variations in traits and how selection could act on those traits. All human experience preceding Darwin said that fish don’t spontaneously spawn spiders etc, so throwing this kind of thing in the mix just muddies the waters. Darwin didn’t need to know the precise mechanism of inheritance in order to draw reasonable broad conclusions from the data he did have available.

Most of the rest of the chapter just gives an overview of the key discoveries over the years that led to the understanding in the 1950’s of the structure of the molecule of inheritance: DNA. Beginning with Mendel’s pea crosses, through observations of chromosomal segregation during cell division, the work of Griffith, Avery, MacLeod, McCarty, Chagraff, Hershey, Chase, and finally Watson and Crick.

I actually think this description is quite well done, as it takes the reader through the discoveries in chronological order, showing how questions were asked by one set of results and then answered by a subsequent experiment.

Jeanson’s point here is that since traits are defined genetically, and since species are defined by traits, the origin of species is fundamentally a genetic question. The punchline is basically: ‘We didn’t uncover the physical basis for heredity until 100 years after Darwin wrote On the Origin of Species, so how could Darwin have had any hope of getting evolution right?’ Despite the fact that Darwin was ignorant of the nature of inheritance, he was still able to infer his conclusions from observations of how living organisms changed over successive generations and inferences about how organisms now preserved in the fossil record changed over geological history. The question of the origin of species requires genetics to provide a complete explanation, but that’s doesn’t mean that the question can’t be partially answered. If someone asked me to explain how a car engine works, I would only be able to explain it at a very shallow level. Not understanding the minute engineering details or combustion chemistry of the process doesn’t negate the broad facts that I do know. Of course, Darwin only explained the process of evolution in very broad terms, that’s why modern Evolutionary Biologists continue to do research to understand the details.

The history of genetics poses a second set of questions to Darwin. Not only was his question a fundamentally genetic one, but his specific answers to the origin of species were also deeply tied to this field. For example, Darwin proposed that all species had one or a few common ancestors. In other words, he said that the vast diversity of life belongs to one or a few family trees. Genealogical relationships are directly recorded in genetics — and nowhere else.

The key word in this passage is “directly”. Of course genealogical relationships are not directly recorded in fossils or the morphology of organisms, but they provide characters that can be used to make inferences about genealogical relationships with some level of confidence. Jeanson is aware of this, he mentioned it earlier in the chapter. Again, Jeanson is acting as though everything has to be perfectly clear-cut when in reality science is based on degrees of evidence. We’ll have to see to what extent Jeanson accepts our ability to reconstruct genealogies though genetics later in the book.

Furthermore, Darwin claimed that new species arose via the process of survival of the fittest, or natural selection. Natural selection is useful to evolution if — and only if — the survivors pass on their superior traits to offspring. In other words, the mechanism of evolution is inextricably tied to inheritance. Inheritance is directly recorded only in genetics.

This statement is a bit odd. It seems to imply that since Darwin didn’t know about genetics, he shouldn’t have postulated natural selection. You don’t need to know about genetics to observe that traits that are selected for end up being passed on to the next generation. For this, Darwin frequently referenced observations from the artificial selection of animals, as will be discussed in future chapters.

Finally, Darwin placed the origin of species on a very long timescale. However, the process of inheritance also acts like a timekeeper, independently recording the length of time over which species appeared (a concept we’ll explore in detail in later chapters). How could Darwin have written On the Origin of Species without any genetic data to test his ideas? Since both his question and his hypotheses were deeply tied to inheritance, what prompted him, not only to pen, but also to vigorously argue for his proposal?

This question will also be answered in later chapters, but some of the reasons Darwin places the origin of species on long timescales should be immediately obvious – species today are observed to change slowly in the wild, and the ancient fossil record suggested that long periods of time were involved, although specifically how much time wasn’t nailed down until radiometric dating came along to really quantify geological ages. The fact that Darwin didn’t have molecular clocks to test his hypotheses doesn’t make them baseless.

Chapter 3 – Cracking the Code

Now that the identity of the molecule of heredity has been identified as DNA and had its structure (double helix) and form in the cell (chromosomes) established, Jeanson moves onto the question of how exactly DNA produces phenotypes.

He begins this story by explaining how the genetic code – how the amino acids in proteins map to the DNA – was discovered. This involved both the codon structure of protein-coding sequences and the “Central Dogma” of molecular biology: DNA being transcribed into RNA and the translation of RNA to proteins (strings of amino acids) by ribosomes and tRNAs.

He then notes that genome size doesn’t correlate well with organismal complexity: lungfish have 20x as much DNA in their genome as humans, and so on. This, and the fact that not all genes appeared to contribute equally to phenotype (e.g. Hox genes play a massive role relative to most genes), are presented to hammer home the fact that there isn’t a simple relationship between the “blueprint of life” (DNA) and organismic complexity. Then he covers a brief history of genome sequencing that culminated in the human genome sequence in 2001. He notes that there didn’t seem to be a correlation between the number of protein-coding genes discovered in various species and their genome size For example, humans have about “20,000-30,000” protein-coding genes in our 3 Gb (3 billion base-pairs) genome, while the 120 Mb (120 million base-pairs) genome of the fruit fly contained almost as many protein-coding genes: 13,600.

Some species’ genomes contain vast amounts of non-coding DNA, so how much of the genome is actually functional? Inevitably, this is where Jeanson refers to the ENCODE project and says that they found “biochemical evidence for function” in 80% of the genome, with the expectation that this figure will rise to 100%:

One of the ENCODE project researchers speculated that, eventually, evidence would accumulate for function in nearly 100% of the human genome.39
This expectation is plausible for at least two reasons.

Paraphrased, those 2 reasons are:

1. The amount of non-coding DNA correlates with organismal complexity better than the amount of protein-coding DNA does.

This doesn’t logically provide evidence that all of the non-coding DNA is functional, just that a relatively consistent proportion might be. For example, if 10% of non-coding DNA was functional in every organism, and that functionality was key to organismal complexity, the amount of non-coding DNA would still correlate very well with organismal complexity. Jeanson also doesn’t take into account the reverse: what if organismal complexity results in the accumulation of a larger amount of non-coding DNA? Larger, more complex organisms tend to have much smaller population sizes and produce fewer offspring than simpler microbial populations – what if more of a buffer is needed in the genome for such populations to survive? That “buffer” could be called “functional” in one sense, but not in the sense that Jeanson has used it so far and will use it later in the book.

Jeanson also failed to note that in the study he’s citing (Liu et al. 2013), “organismic complexity” is measured in terms of the number of cell types, which is relatively constant in groups like mammals, for example. Instead, he leaves the definition of “more complex” up to his readers’ imaginations. Since most of these people will be creationists, will they intuitively know that Jeanson is thinking about complexity in terms of the number of cell types, or will they think about complexity in a very different way, such that humans are much more complex than all other mammals, for instance? The proportion of the genome that is non-coding is also pretty consistent, ranging from 98.5% to 99.1% in mammals in their study. In other words, non-coding DNA proportions can correlate with the number of different cell types in a species at a very zoomed-out scale – more like the phylum level in animals.

2. ENCODE didn’t survey the genome during developmental stages, so more “biochemical evidence of function” is yet to be discovered (in those stages).

By making this point, it becomes clear that Jeanson is specifically answering the question of “how can we get from 80% functional according to ENCODE to 100% according to ENCODE?”, while completely skipping over the question of “does the biochemical activity that ENCODE has catalogued actually translate to function?” Jeanson is aware that this is the most controversial point surrounding ENCODE’s results, but he doesn’t even give it a passing mention in this chapter. He concludes:

Thus, when examined historically, the evidence for genome-wide function is gaining strength.
Together, these experiments suggested that the majority — if not the vast majority — of gene and non-genic DNA sequences were functional.

I don’t want to get bogged down on the subject of ENCODE here since there’s a huge amount to talk about if you want to cover it in real depth, but in a nutshell, ENCODE’s analysis didn’t look at “biochemical function” it looked at “biochemical activity“. An oft-used analogy is that chewing gum on the sidewalk can have the activity of sticking to your shoe, but that doesn’t mean its function is to stick to your shoe. ENCODE surveyed the genome for lots of different kinds of biochemical activity such as protein binding and being transcribed, but these aren’t synonymous with functionality. It was known long before ENCODE that the majority of the genome was transcribed, for example, but this is likely mostly just “transcriptional noise”. For a more on ENCODE’s results and how they relate to function, I highly recommend reading this paper which explains it in fairly simple terms: Graur et al. 2013. The important point is that ENCODE has not demonstrated that 80% of the human genome is functional (the real number seems to lie between ~10-15% under most definitions of “functional”), just that most of the genome interacts with other biological molecules in some way, something that, while very useful to understand the details of, isn’t particularly surprising. More on ENCODE coming later in this series.

Jeanson concludes this chapter by continuing to narrow in on what it takes to really the answer of the origin of species:

If we want to understand the origin of species, we must uncover the origin of the first traits. Since traits are ultimately encoded by DNA, the origin of species is a question of the origin of DNA differences within and between species.

In other words, the answer to the origin of species can be uncovered for the first time right now.

It’s true to say that if we really want to thoroughly understand the process of how new species arise, we have to look at the DNA. Given that these are literally the last sentences of the chapter and of the first Part of the book, you might expect the narrative to continue along the lines of DNA and genetic differences, but no, the second part (the next 3 chapters) of the book go back to Darwin, and how he supported his conclusions about evolution without reference to DNA.

Aside from the gripe that I have with this chapter about non-coding DNA, it’s another fairly mundane one – taking 58 epub pages (~24 physical pages) to describe in very basic terms how DNA relates to phenotype.  I think it could have done with a decent description of gene structure and how strings of amino acids actually end up as function proteins (i.e. folding), but this isn’t really necessary, because despite what this chapter hints at, the rest of the book won’t be about the details of the molecular evolution of functional elements.

I was planning to do this series chapter-by-chapter, but there wasn’t too much to comment in these 2 chapters because they were just explaining the basics of inheritance and DNA to a lay audience, so I decided to bundle these together as a bumper pack. Now that these foundations have been laid, the future chapters will contain a bit more content (with the possible exception of Chapter 4).

Comments and queries are welcome.



8 thoughts on “Reviewing “Replacing Darwin” – Part 2: Darwin Didn’t Know About Chromosomes!

  1. Many thanks for your review of chapter 3 of Jeanson’s book. I’ve not got the book or read it, BTW, just followed it on some youtube videos. I’m also just a layman fascinated by and trying to follow the arguments.

    I’m not entirely sure I understand your critique of Jeanson’s reasoning #1 and #2.

    Regarding #1 – Isn’t Jeanson’s point that there is speculation, by at least one ENCODE project researcher, that we may discover (closer to) 100% functionality in the genome? In other words, right now, we don’t know what we don’t know? Therefore, speculating that only 10% of non-coding DNA has functionality, at this point in time, is quite arbitrary and premature. Likewise speculating about “a buffer is needed in the genome for such populations to survive” is just that – pure speculation. Why would Jeanson be negligent in failing to hypothesize about buffers, when he could equally speculate a hundred other alternatives? But he would be right in speculating the opposite, because that’s what the trend seems to be; for it was thought that non-coding DNA was 100% junk at one time, but that proportion is shrinking, it seems, day by day. And why would you even want to slap a 10% limit on it, when the whole thing is so complex that we’re discovering new things every day?

    Regarding #2 – I kind of get what you mean. But I think we’re back to the junk-DNA analogy again. The argument seems to be that just because there is resultant biochemical activity does not mean there is actual biochemical functionality. I understand the point. But aren’t we right at the beginning of such research? There may well be new and exciting discoveries to be made. I’m so grateful that scientists did not abandon, or could not abandon, delving into a supposed pile of junk-DNA because we are now discovering rich pickings there. I hope that conscientious scientists will likewise not dismiss biochemical activities as mere junk of little consequence, but will do due diligence to ascertain whether these activities correlate to subtle/amazing functionalities yet to be uncovered.


    1. Thanks for your comment.

      Yes, Jeanson mentions several times that there’s speculation that up to 100% of the genome will be found to be “functional”. What he doesn’t even mention is that since those speculations come from ENCODE researchers, they’re using the ENCODE definition of “function”, which is basically “activity”. In other words, he’s telling his audience that researchers in the field are saying one thing, when they’re actually referring to something else. I would agree that essentially 100% of our genome will eventually be found to have “biochemical activity”, but this is *very* different from function. It’s not like we’re saying “well there’s activity, but we don’t *really* know if it’s functional so let’s not call it functional just in case”, it’s more like “of course there’s activity, that’s what we’d expect of any genome, but that’s completely different from establishing functionality”. “Activity” and functionality are worlds apart. Activity is required for functionality but that’s like saying “mass is required for something to be edible” – it doesn’t mean that everything is edible, not even close. It would be ridiculous for anyone to conclude that everything with mass was edible just because it’s established that mass is a property required for food. It’s a fallacy calling “affirming the consequent”.

      “it was thought that non-coding DNA was 100% junk at one time”
      Yes… At one time. We’ve know that at least some of the non-coding DNA was highly relevant to function since the 1950s, and it’s been widely accepted that non-coding sequences play an important role in the evolution of different functions between species since the 1970s. The proportion of DNA considered to be non-functional is shrinking day by day, but that’s because that’s pretty much the only direction our knowledge can go! Going from a position of being ignorant about basically all functionality in the genome, say, a century ago, documented cases of function can only increase. The only way the number would reverse would be if previous results were found to be fundamentally flawed, which happens much less. That doesn’t mean it’s reasonable to infer that the number is tending to 0%. For example, it could be asymptotically approaching a number like 80%.

      As I said, I didn’t want to go into much detail about this in the blog post because this subject is so full of things to unpack, and that’s why I didn’t explicitly clarify that we’re not basing our estimates on the proportion of our genome that is functional purely on a lack of evidence for function in most of it. What I said about a “buffer” wasn’t pure speculation at all, it is a statement based on evidence from research for more than 40 years on population genetics. One of the most key reasons we’re pretty confident that only around 15% of the genome is functional (in terms of requiring a specific sequence, which is the only way it could be relevant to creationist arguments), is because of this research into genetic load. I recommend you read the 2013 article by Graur et al. that I linked in the post (it’s quite accessible for laymen), and you could supplement that by reading the 2014 article by Palazzo and Gregory called “The Case for Junk DNA”. For a more technical look at the mathematics of the argument I’d recommend the 2017 article “An Upper Limit on the Functional Fraction of the Human Genome”, also by Dan Graur. The take-home point here is that we’re absolutely not basing our numbers purely on argument about a lack of evidence, they’re also based on positive evidence that explicitly tells us only a small fraction of the genome *can possibly be* functional.

      Don’t worry, many scientists are absolutely working on discovering new functionalities in our genome, and the genomes of other species. Non-coding DNA does indeed contain a treasure-trove of information (that’s why I work on it), but while we don’t know all of the contents of the trove yet, we do have a pretty good idea about the size of it.

      Liked by 1 person

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s